Detection: Linux Auditd Shell Execution by Non-User Account

Description

The following analytic detects successful auditd execve events that launch an interactive Linux shell, identified by a TTY, under accounts that commonly do not log in. These are identified by a list of well-known service account names or by a system UID (between 1 and 999). An interactive shell for these accounts may indicate password-bypass abuse or unauthorized access.

 1`linux_auditd` type=SYSCALL
 2comm IN (
 3    "ash",
 4    "bash",
 5    "csh",
 6    "dash",
 7    "fish",
 8    "ksh",
 9    "mksh",
10    "rbash",
11    "sh",
12    "tcsh",
13    "yash",
14    "zsh"
15)
16success=yes
17
18
19| where isnotnull(tty) AND tty!="(none)"
20
21
22| rex field=_raw "\\bUID=[[:punct:]]?(?<audit_user>[[:alnum:]_.-]+)"
23
24
25| eval audit_user=lower(coalesce(audit_user, user))
26
27| eval uid_num=tonumber(uid)
28
29
30| where (uid_num>0 AND uid_num<1000) OR in(audit_user, "daemon", "bin", "sys", "sync", "shutdown", "halt", "games", "man", "lp", "mail", "news", "uucp", "operator", "www-data", "backup", "list", "irc", "gnats", "nobody", "_apt", "systemd-network", "systemd-resolve", "messagebus", "uuidd", "sshd", "ntp", "ftp", "chrony", "postfix", "redis", "mysql", "mariadb", "postgres", "alpine", "guest")
31
32
33| rename host as dest
34
35
36| stats count min(_time) as firstTime
37              max(_time) as lastTime
38              values(audit_user) as user
39              values(uid) as uid
40              values(euid) as euid
41              values(comm) as shell
42              values(exe) as executable
43              values(tty) as tty
44  by dest
45
46
47| `security_content_ctime(firstTime)`
48
49| `security_content_ctime(lastTime)`
50
51| `linux_auditd_shell_execution_by_non_user_account_filter`

Data Source

Name Platform Sourcetype Source
Linux Auditd Syscall Linux icon Linux 'auditd' 'auditd'

Macros Used

Name Value
security_content_ctime convert timeformat="%Y-%m-%dT%H:%M:%S" ctime($field$)
linux_auditd_shell_execution_by_non_user_account_filter search *
linux_auditd_shell_execution_by_non_user_account_filter is an empty macro by default. It allows the user to filter out any results (false positives) without editing the SPL.

Annotations

- MITRE ATT&CK
+ Kill Chain Phases
+ NIST
+ CIS
- Threat Actors
ID Technique Tactic
T1059.004 Unix Shell Execution
T1078.003 Local Accounts Initial Access
Exploitation
Delivery
Installation
DE.AE
CIS 10

CVE

Default Configuration

This detection is configured by default in Splunk Enterprise Security to run with the following settings:

Setting Value
Disabled true
Cron Schedule 0 * * * *
Earliest Time -70m@m
Latest Time -10m@m
Schedule Window auto
Creates Finding (Notable) No
Creates Intermediate Finding (Risk Event) Yes
Anomaly detections generate Intermediate Findings (Risk Events). They do not generate a Finding (Notable) directly.

Implementation

To implement this detection, you should ingest auditd data, that consists of SYSCALL, TYPE, EXECVE and PROCTITLE events, which captures command-line executions and process details on Unix/Linux systems. These logs should be ingested and processed using Splunk Add-on for Unix and Linux (https://splunkbase.splunk.com/app/833), which is essential for correctly parsing and categorizing the data. The next step involves normalizing the field names to match the field names set by the Splunk Common Information Model (CIM) to ensure consistency across different data sources and enhance the efficiency of data modeling. This approach enables effective monitoring and detection of Linux endpoints where auditd is deployed.

Known False Positives

System administrators and automation may run shells under service accounts or accounts that are not locked on a particular image. The search also matches any account with a UID between 1 and 999 (the system account range on most modern distributions), so administrators who use su or sudo to open an interactive shell as a service account such as jenkins or gitlab-runner will trigger it. On legacy systems (for example RHEL/CentOS 6) regular users may start at UID 500, so adjust the threshold if needed. Review the account's shadow state and the process context before escalating.

Associated Analytic Story

Intermediate Findings

Message Entity Field Entity Type Risk Score
Shell [$shell$] executed as non-user account [$user$] on [$dest$]. dest system 20

References

Detection Testing

Test Type Status Dataset Source Sourcetype
Validation ✅ Passing N/A N/A N/A
Unit ✅ Passing Dataset auditd auditd
Integration ✅ Passing Dataset auditd auditd

Replay any dataset to Splunk Enterprise by using our replay.py tool or the UI. Alternatively you can replay a dataset into a Splunk Attack Range


Source: GitHub |

Version: 1